Solid-phase synthesis of Avian β-Defensin 8
Author: Erik Selim Supervisor: Håkan Andersson Examiner: Kjell Edman Semester: HT 13 Subject: Chemistry
Abstract
Differences in the expression of antimicrobial peptides in vivo have been proposed as underlying factors influencing susceptibility to infection. In this context, the role of avian β- defensins in inhibiting avian influenza infections is a study object in an ongoing collaboration with the Zoonotic Ecology and Epidemiology group at Lnu.
In this report, an attempt to synthesize two variants of the peptide Anas Platyrhynchos AvBD-8, using Fmoc-based SPPS, is described. The length of AvBD-8 (43 aa) necessitated peptide synthesis in two segments to subsequently be ligated using native chemical ligation. The first component of a 19 aa segment was thus a Dbz-linker, which would allow to ligate this end with a second segment (24 aa). Halfway through the synthesis of this larger segment the batch was split into two pots, allowing the synthesis of two segments differing by one single amino acid (R for W). The composition of these segments were: Dbz-HDTSCTGGAQKCQVANNPA (Dbz- segment), SVVTRCCPIGQKCWGFARTNPPPC(boc) (W-segment), and
SVVTRCCPIGQKCRGFARTNPPPC(boc) (R-segment). Crude product yields were 284,5 mg;
67,6% (Dbz-segment), 137,6 mg; 52,3% (W-segment), and 166,3 mg; 64,2%. Preliminary mass spectrometric analysis on the crude products did not indicate the presence of the desired
segments in major mass peaks. Further product purification is necessary in order to allow definite conclusions, but it appears as if the synthesis has not worked. Possible explanations are either impure or degraded reactant(-s), folding or shielding effects of the growing peptide chain at some point inhibiting synthesis, or experimental errors during one or more of the many steps involved in the synthesis.
Keywords
SPPS-Peptide Synthesis-Defensin-Mallard duck-Avian Beta-AvBD8
Acknowledgements
I would like to acknowledge my supervisor Dr Håkan Andersson, Dr Minh-Dao Duong- Thi for help with mass spectrometry, and Dr Joanne Chapman for a generous supply of literature.
Sammanfattning
Skillnader i uttryck av antimikrobiella peptider in vivo har förslagits kunna påverka infektionskänslighet. I detta sammanhang är den inhiberande rollen av fjäderfäns beta- defensiner har mot influensainfektioner ett forskningsobjekt i en pågående samverkan med zoonosgruppen på LNU.
I denna rapport beskrivs ett försök att syntetisera två varianter av peptiden Anas Platyrhynchos AvBD-8 genom att använda en variant av metoden SPPS som är Fmoc-baserad. Längden på AvBD-8 (43 aminosyror) krävde att peptiden syntetiserades i två segment för att därefter ligeras (NCL-metoden). Den första komponenten av ett 19 aminosyror långt segment var således en Dbz-länk som skulle möjliggöra ligeringen mellan denna ände och ett andra segment med en längd av 24 aminosyror. Halvvägs genom syntesen av det större segmentet delades batchen upp i två kärl för att möjliggöra syntes av två varianter, vilka endast skilde sig m.a.p. en aminosyra (R och W). Sammansättningen av dessa segment var: (Dbz-segmentet) Dbz-
HDTSCTGGAQKCQVANNPA, (W-segmentet) SVVTRCCPIGQKCWGFARTNPPPC(boc) och (R-segmentet) SVVTRCCPIGQKCRGFARTNPPPC(boc). Råproduktsutbytet var 137,6 mg (52,3%) för W-segmentet, 166,3 mg (64,2%) för R-segmentet och 284,5 mg (67,6%) för Dbz- segmentet. Den preliminära masspektrometriska analysen av råprodukten gav inga indikationer på att de önskade segmenten var närvarande i de större masstopparna. Vidare är upprening av produkten nödvändig för att kunna dra några definitiva slutsatser, men det verkar som att syntesen inte har fungerat. Möjliga förklaringar är antingen oren(a) eller degraderad(e) reaktant(er), veckning eller skyddande effekter av en växande peptidkedja som vid någon tidpunkt börjat inhibera syntesen, eller experimentella misstag vid ett eller flera av de många steg som var involverade i syntesen.
Content
1 Introduction _________________________________________________________ 1 1.1 Objectives _______________________________________________________ 1 1.2 Antimicrobial peptides _____________________________________________ 1 1.3 Methods for peptide synthesis _______________________________________ 3 1.4 Solid-phase peptide synthesis ________________________________________ 3 1.5 Ninhydrin test ____________________________________________________ 5 1.6 Fmoc thioester linkage and native chemical ligation ______________________ 5 1.7 LC-MS _________________________________________________________ 6 1.8 Freeze-drying ____________________________________________________ 8 2 Materials ____________________________________________________________ 9 3 Method _____________________________________________________________ 9 4 Results _____________________________________________________________ 11 5 Discussion __________________________________________________________ 14 6 References__________________________________________________________ 16
1 Introduction
1.1 Objectives
The main objective of this work was to synthesize two variants of Anas Platyrhynchos avian β-defensin-8 (AvBD-8):
APNNAQCKQAGGRCSTDHCPPPNTRAFGWCKQGIPCCRTVVS and
APNNAQCKQAGGRCSTDHCPPPNTRAFGRCKQGIPCCRTVVS using an Fmoc- SPPS approach. The peptides should be synthesized in two segments with the lengths of 24 and 19 amino acids, respectively. One of the segments was to be synthesized with the incorporation of a DBZ-linker (diamino benzoic acid-linker) and the synthesis of the other segment divided into two batches halfway through the synthesis, to obtain two separate segments where one amino acid is replaced with another to achieve the two AvBD-8 alleles. After completed segment synthesis, purification and structure
verification of the three segments should be carried out through LC-MS. This was to be followed by joining these segments into the two AvBD-8 variants through native chemical ligation (1). Finally, to obtain the proper folding of the peptides and to create the appropriate disulfide bridges the ligated defensins were to be oxidized. The
structures of the folded, oxidized peptides were to be confirmed by NMR, and subjected to analysis of their antimicrobial activity.
1.2 Antimicrobial peptides
Antimicrobial peptides (AMPs) are small polypeptides containing less than 100 amino acids, which exhibit antimicrobial activity in their natural environment and at
physiological concentrations (1). The AMPs are a part of the immune system and most of them share some common features, i.e. an amphipathic structure or containing a positive net charge (2). Since at least 1200 AMPs have been identified or predicted, several categories of AMPs have been established in concordance with their amino acid compositions, conformational structures and sizes (2).This categorization allows the AMPs to be divided into different groups, i.e. peptides with α-helix structure, peptides with loop structure or peptides with β-sheet structures which are stabilized by disulfide bridges (2).
The expression and induction of AMPs at the epithelial surface have been shown to have an effect against a number of different infectious agents such as bacteria, fungi, viruses and parasites (2). The expression of AMPs can be very different depending on from which tissue and cell type it originates, but more often than not a number of AMPs are co-expressed as a group, acting in concert (2). One example of this co-expression can be found in the skin where 20 different AMPs and proteins have been identified, including two groups of AMPs which have been studied extensively (2). These two groups are called defensins and cathelicidins and have been found in mammalians, as well as in other vertebrates (2, 3). All cathelicidins have a highly conserved domain near the N-terminal called cathelin, which is also what their name is based on (2). The name defensins on the other hand originates from their association with the host immune system (1). Defensins are also the most expressed AMPs in humans, and a wide variety is present in infected or inflammated human tissue (1).
The common characteristics of defensins are the six cysteines linked together in pairs by disulfide bridges and the folding of the peptides, dominated by the high prevalence of β- sheets (1). Further characteristics shared by defensins are that they are cationic,
microbicidal and lacking glycosyl- or acyl-side chain modifications (2). Defensins can also be described as a family of evolutionarily related vertebrate AMPs (1). Some of the
defensins exhibit broad antimicrobial activity and they make up an important part of the innate immune system (4). The defensins are believed to utilize their hydrophilic and positive electrostatic properties to bind negatively charged membranes of pathogens, thereby disrupting membrane structure to cause lysis (4). Defensins might also, in addition to their antimicrobial properties, contribute to the immune system by mediating signalling to macrophages, lymphocytes and mast cells (4). The defensins are divided into three subfamilies depending on the connectivity between the cysteine residues and the degree of sequence homology (2). The subfamilies are called α-, β- and θ-defensins, where α- and β-defensins are the two main families (1, 2). The differences between α- and β-defensins consists in the distances between the cysteine residues and the linking of these cysteines (1, 2). In α-defensins (illustrated below in fig.1) the linking pattern is 1-6, 2-4 and 3-5 whereas the linking pattern of β-defensins (illustrated below in fig. 1) is 1-5, 2-4 and 3-6 (2). The θ-defensins consists of two hemi α-defensins, which together form a cyclic structure without a free N- or C-terminus (1, 2). Further the θ-defensins are inactivated in humans due to mutations that encode premature stop codons (1).
Figure 1: Illustration of an alfa-defensin (left) and a beta-defensin (right).
( http://www.ebi.ac.uk/pdbe-srv/view/entry/1ews/summary&http://www.ebi.ac.uk/pdbe- srv/view/entry/1bnb/summary)
An ancestral β-defensin gene is believed to be the precursor of both α- and θ-defensins due to the fact that only β-defensins have been discovered in phylogentically older vertebrates i.e. birds and fish (3). Furthermore, it is hypothesized that all β-defensins originate from a single ancestral gene (3). β-defensins were initially discovered in epithelial cells and neutrophils from cattle (5). Since their discovery, they have been isolated from almost all the vertebrates that have been studied (5). In humans at least 5 gene clusters of defensins have been found, which all correspond to syntenic mouse clusters (1). The structural parameters (i.e. primary structure, charge, hydrophobicity) of β-defensins play an important role for the interaction with the microbial cell wall (6).
These structural parameters determine the differences between β-defensin specificities in terms of in vitro potency, gram specificity and strain specificity (6). However β- defensins have been shown to also contribute to the innate immune system by exerting chemotactic activity on dendritic cells, T-cells, monocytes, macrophages and mast cells (6). Moreover, some β-defensins have demonstrated an inhibiting effect in vitro on HIV-1 replication (6). Large amounts of expressed β-defensins have been observed in the epididymis (6).
As previously mentioned, β-defensins have been found in many vertebrates, i.e. fish and birds (3). β-defensins appear to be the only defensins expressed in avian species and are designated Avian β-defensins (AvBDs) (7). These have been found in many avian species, including chicken, goose, quail, king penguin, ostrich, king pigeon, turkey and duck (3, 7). Similar to other β-defensins, the AvBDs exhibit activity against bacteria and fungi (7). Most of the AvBDs can either be induced due to microbial infection, or expressed constitutively (7). In the case of ducks, currently at least 11 β-defensins have been identified. These birds evoke a medicinal interest due their frequent exposure to virus infections (7). The medicinal interest lies in the experimental possibilities of antimicrobial treatment in a time when antibiotic-resistant bacteria is on the rise (7).
Duck hepatitis virus (DHV) infection is a relatively common condition, which has been used to examine the antiviral activity of some AvBDs from ducks (7). Some of the defensins acquired from ducks indeed exhibited such activity (7).
Further an interesting aspect of the defensins (including duck defensins) are their variability in sequence due to mutations (1, 2). Early speculations have been made considering if these mutations are an evolutionary response to cope with the altering microflora in animals of different species (1, 2). These specualtions have found support by studies comparing the homology of amino acid sequences between two species i.e.
duck AvBD-4 and chicken AvBD-4 shared 71,9 % amino acid homology (8). The interest further lies in the identification of effects that these alterations may have on antimicrobial activity and how the differentiation in amino acid compositions effects the specificity of microbial targets (1).
1.3 Methods for peptide synthesis
Since the first report of dipeptide synthesis in 1901 by Emil Fischer, chemical peptide synthesis has attained growing interest throughout the world (9, 10). Today two techniques dominate chemical peptide synthesis: solution phase synthesis (SPS) and solid phase peptide synthesis (SPPS) (9). The coupling of single amino acids in solution is called SPS (9). To form longer peptides with SPS, the most common strategy is a
“convergent synthesis” (9, 11). The peptide is built by assembling small parts of the peptide, and upon isolation and purification the fragments are coupled into a full sequence peptide (11). One advantage with SPS is the possibility to purify intermediate products to give the final peptide a higher purity (9). A disadvantage of SPS are the long reaction times, a major reason for using SPPS, although large-scale SPS, in comparison, is quite inexpensive (9). The usage of protecting groups is quite common in SPS, with Boc (tert-butyloxycarbonyl) protection (illustrated below in fig.2) as the industry’s first choice (12). SPPS is the assembling of amino acids onto a solid matrix, which allows for easy washing steps. This strategy will be described in detail below (9, 13). Further progress with regard to the SPPS strategy involves microwave-assisted SPPS where the irradiation from microwaves helps to obtain a higher yield and a lower degree of racemization (9). Both regular and microwave-assisted SPPS are carried out with protecting groups (9).
Figure 2: The Boc group coupled to an amino acid
1.4 Solid-phase peptide synthesis
The growing interest in protein chemistry in the early 20th century resulted in scientific progress of peptide synthesis where development of amino-protecting groups like the carbobenzoxy (Cbz) group in 1931 and the tert-butyloxycarbonyl (Boc) group in 1957 were important discoveries (9). A major breakthrough in peptide chemistry came with Bruce Merrifield’s development of solid-phase peptide synthesis (SPPS) in 1963, for which he later, in 1984, received the Nobel prize in chemistry (9), (13). Although previous methods had allowed to create complex products such as oxytocin and porcine gastrin releasing peptide, their synthesis was time consuming and difficult (13). A new approach was needed for more complex peptides to be synthesized (13). The idea was to couple the amino acids in a stepwise manner with a solid support, which would make
the peptide insoluble and allow rapid washing and filtration (13). The amino acids added to the insoluble matrix has amino-protecting groups, i.e. when the first amino acid is attached, both the reactive side chain and the alfa-amino group is protected to ensure that the bond is generated between the C-terminal and the resin (9). Upon completion byproducts are washed away and the next coupling sequence commences with deprotection of the protecting group of the α-amino group (9). Two popular
amino-protecting groups are Boc and Fluorenylmethyloxycarbonyl (Fmoc) (9). The Boc groups are removed by 50 % trifluoroacetic acid (TFA) in dichloromethane (9, 13). The peptide is subsequently exposed to a tertiary amine i.e. diisopropylethylamine (DIPEA) to neutralize the α-amine salt, which enables the coupling of the next Boc-protected amino acid (9, 13). Upon completion of the synthesis the resin is separated from the resin by cleavage with a strong anhydrous acid, i.e. hydrofluoric acid (HF) (9, 13).
Before initiation of the synthesis the resin is treated with organic solvents i.e.
dichloromethane and dimethylformamide (DMF) to allow the resin to swell and increase the reaction rates (9, 13). The Fmoc groups are removed with the aid of 20 % piperidine in DMF (9, 13). The neutralization of the α-amine salt is carried out in the same way as for the Boc-groups, with a tertiary amine i.e. DIPEA (9). A basic schematic of SPPS using a Fmoc approach is illustrated below in fig. 3.
Figure 3: Schematic illustration of SPPS using Fmoc protecting groups. The protecting groups of the amino acid side chains are not illustrated.
1.5 Ninhydrin test
The ninhydrin test, also known as the Kaiser test, is based on the reaction between heated ninhydrin (triketohydrindene hydrate) and a free amino acid α-amine under acidic conditions (14). The products consist of ammonia, carbon dioxide and a blue- purple complex (14). The blue-purple complex is utilized when quantifying the amount of free amines by measuring the absorption at its maximum (570 nm), which enables the calculation of the coupling yield in SPPS (13, 14). Other amines present will also react with ninhydrin but since these are protected in SPPS chemistry they will not participate in the reaction (13, 14). The exception is proline which has a free secondary amine and therefore is ineligible for the ninhydrin test (13, 14).
1.6 Fmoc thioester linkage and native chemical ligation
One of the more popular approaches when ligating two unprotected peptide segments through a native peptide bond is native chemical ligation (NCL) reaction (15). NCL can be utilized both for synthesizing circular peptides and proteins as well as long difficult peptides (15). The requirements for the ligation is a C-terminal α-thioester and an N- terminal cysteine (15). When synthesizing peptides with Boc-SPPS the NCL can be initiated immediately since the thioesters are stable to the acidic deprotection used (15).
Using NCL in Fmoc-SPPS can be somewhat problematic since the deprotecting agent of Fmoc, piperidine, is incompatible with the thioesters due to the nucleophilicity of piperidine (15). However, a method has recently been developed for generation of thioesters that are compatible with Fmoc-SPPS (16). The discovery of the stable synthetic 0-aminoanilide intermediates (illustration of an 0-aminoanilide intermediate below in fig.4) made way for this method, which is based on C-terminal N-acylurea peptide mediety (15, 16). Upon completion of the peptide assembly the 0-aminoanilides are transformed to generate an N-acylbenzimidazolinone (N-acylurea) by undergoing acylation and cyclization (15, 16). The N-acylurea (Nbz) is stable in acidic conditions, but in neutral aqueous conditions the unprotected Nbz-peptide reacts and undergoes rapid thiolysis (15, 16). The thiolysis creates the C-terminal α-thioester which is, as described above, necessary for the NCL (15, 16). The NCL mechanism (which is illustrated in fig.5 below) is based on the reversible reaction where the thiolate group of the N-terminal cysteine residue attacks the C-terminal thioester leading to the formation of an intermediate (15). The ligation of the two peptide parts is completed by an
intramolecular S,N-acyl shift (15).
Figure 4: Illustration of an 0-aminoanilide intermediate (Dbz-linker) coupled to an amino acid.
Figure 5: The mechanism of NCL where the thiolate group of the N-terminal cysteine residue attacks the C-terminal thioester and ligation later on is completed by an intramolecular S,N-acyl shift. (http://en.wikipedia.org/wiki/File:NCL_mechanism.pdf Author: AdLucem)
1.7 LC-MS
LC-MS is a combined separation and detection technique (17). The basic setup of liquid chromatography separation is a system where solvent is continuously pumped from a reservoir through a column (17). The samples of interest are added into the column through an injection valve (17). The passage through the column separates the sample molecules to be detected later on (17). The column is a tube with cylindrical shape which is filled with spherical-shaped particles (17). The diameter of these particles are usually between 1,5 – 5 µm (17). The particles are often of porous silica with an attached stationary phase i.e. C18 groups (17). Every spherical particle consists of aggregated subparticles (17). The spaces between the subparticles create the pores in which the molecules are delayed due to the extra distance (17). The mobile phase flows through the column together with the sample molecules, which diffuse into the particles (17). There are different setups within liquid chromatography where normal-phase and reverse-phase are two common approaches (17). In reverse-phase chromatography (RPC) a nonpolar column is used i.e. C18 (which is the most common one and therefore the one described) and the mobile phase is a polar blend of water and organic solvent i.e. acetonitrile (17). The normal-phase chromatography (NPC) utilizes a setup where the column is polar i.e. unbonded silica and the mobile phase consists of less polar organic solvents i.e. hexane and methylene chloride (17). Together with RPC and NPC there are various other chromatographic modes which can be employed such as ion- exchange chromatography (IEC), ion-pair chromatography (IPC), size-exclusion chromatography (SEC), hydrophilic interaction chromatography (HILIC) and non- aqueous reversed-phase chromatography (NARP) (17). Below in fig. 6 a schematic picture of a typical HPLC equipment is illustrated.
Figure 6: The picture is a equipment of a typical HPLC machinery where 1 represents solvent reservoirs, 2 solvent degasser, 3 gradient valve ,4 mixing vessel for delivery of mobile phase, 5 high-pressure pump, 6 switching valve in inject position, 6´ switching valve in load position, 7 sample injection loop, 8 pre-column ,9 analytical column, 10 detector, 11 data acquisition, 12 waste bottle. (http://commons.wikimedia.org/wiki/File:HPLC_apparatus.svg Author:
WYassinemrabetTalk)
The mass spectrometric detector is the most commonly used detector for bioanalytical methods (17, 18). The basic principle of MS is that the molecule(s) of interest is ionized and sent through a mass filter i.e. a quadrupole mass filter where the ions and fragments are sorted after mass (19). The intensity of the different fragment beams and masses are detected and transformed into an electrical signal which is translated into a mass
spectrum with the aid of a computer (19). When dealing with a liquid sample there has to be an ionization process and for LC-MS there are two popular methods, APCI (atmospheric pressure chemical ionization), which is illustrated in fig. 7 and ES
(electron spray) (19). In ES ionization the sample is ionized through a needle along with the solvent (19). In the point of the needle, high voltage is applied and the small
droplets receives a charge due to the lower voltage of the counter-electrode (19). Due to the vaporization of the droplets an increased net charge is obtained which leads to such a big charge that the ions explode (19). This reaction repeats itself until the ions are transformed into gas-phase and proceeds through a capillary tube, towards the mass filter (19). APCI sprays the solvent and sample along with nitrogen to vaporize it into an oven (150-350 °C) (19). The solvent becomes ionized due to collisions with electrons from a corona needle (19). The solvent ions then ionizes the sample by chemical ionization (hence the name) (19). The sample is transported towards the mass filter with a capillary tube which is charged with the opposite of the ion (19). There are different types of mass filters but the most common one used together with LC-MS is the quadrupole mass filter, which is illustrated in fig. 8 (19). The quadrupole consists of four poles which are symmetrical arranged (19). There are two high-pass filters and two low-pass filters, and in a three dimensional image the high-pass filters are aligned in a y-axis and the low-pass filters are aligned in an x-axis (19). When the quadrupole is set on a specific m/z (where m is molecular or atom mass number, z is the ions charge number and the / should be interpreted as ratio) the idea is that ions with the right m/z pass through along the z-axis and the others are removed with the aid of the high- and low-pass filters (19). When the quadrupole is set on scanning the filters are vibrating with a determined variation in frequency to allow ions of different m/z to pass the filter
(19). The ions that pass the filter are detected and transformed into an electrical signal which is sent to a computer (19).
Figure 7: Illustration of the ionization method APCI were liquid is vaporized and thereafter ionized. (http://en.wikipedia.org/wiki/File:Apci.gif Author: Kkmurray)
Figure 8: Illustration of the movement of ions with both correct and incorrect M/Z in a quadrupole mass filter.http://commons.wikimedia.org/wiki/File:Analyseur-quadripolaire- MS.pngAuthor: Gbdivers. Modification: altered text from the original picture)
1.8 Freeze-drying
Freeze-drying is a preservation method commonly utilized for extending the durability of perishable materials and for allowing transport without refrigeration (20). The basic principle of freeze-drying is to reduce the pressure in the surrounding of the frozen product to allow the water in the product to convert from solid phase to gas phase by sublimation (20). The freeze-drying, which is a dehydration process, can be divided into four stages: pre-treatment, freezing, primary drying and secondary drying (20).
All treatment methods prior to freezing are considered to be pre-treatment (20). There are various pre-treatments such as decreased concentration of a high vapor pressure solvent, an increased surface area or addition of components to increase
stability/improve processing (20).
The most critical stage of the freeze-drying is stage two, the freezing, since if performed inadequately the product can be spoiled (20). The common approach in laboratories is to use a shell freezer, which obtains low temperatures by mechanical freezing, dry ice and methanol or liquid nitrogen (20). The product is placed in a freeze-drying flask and subsequently lowered into the shell freezer where the flask is subjected to rotation during the freezing (20). The material’s triple point is the lowest temperature where
both the solid and liquid phase of the material can coexist (20). The aim of the freezing is to obtain a temperature below the triple point of the solvent of the materials (i.e.
water), which enables the solvent to sublimate rather than to melt in the drying
procedures afterwards (20). The temperature in the shell freezer is often in the interval from -50°C – -80°C (20).
In the primary drying phase, pressure is reduced to a few millibars and a sufficient amount of heat is added to enable the sublimation of the solvent (20). The amount of heat needed can be calculated and it is important that an excessive amount of heat is not used since the material structure then may be altered (20). The primary drying
sublimates approximately 95 % of the water from the material (20).
The goal of the secondary drying phase is to remove unfrozen water (20). The mechanism of the process is driven by the adsorption isotherms of the material (20).
The temperature is higher in comparison to the primary drying phase, whereas the pressure can be somewhat lower (20). Upon completion, the vacuum is often broken with an inert gas i.e. nitrogen, and thereafter sealed (20). The final residual water content in the product is in the range of 1-4 %, which is an extremely low concentration (20).
2 Materials
Peptide synthesis
Dimethylformamide (DMF) was purchased at VWR. The FMOC-protected amino acids, Boc-Cys(Trt)-OH and O-(Benzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
hexafluorophosphate (HBTU) were retrieved from Iris Biotech GmbH. Piperidine were Bought from Sigma-Aldrich along with Diisopropylethylamine (DIPEA). DBZ-linker was retrieved from Anaspec Inc and Tentagel R RAM resin was purchased from Peptides international.
The ninhydrin test
Potassium cyanide (KCN), ninhydrin and phenol were purchased from Sigma-Aldrich.
Ethanol was from Solveco.
Peptide cleavage
TFA (Trifluoroacetic acid) was from Iris biotech GmbH and TIPS (triisopropylsilane) was from Sigma-Aldrich. MeCN (Acetonitrile) was obtained from Merck.
Peptide ligation
MPAA (4-nitrophenyl chloroformate) and TCEP (4-mercaptophenylacetic acid) was purchased at Sigma-Aldrich.
3 Method
Peptide synthesis
Prior to synthesis 20% piperidine in DMF (dimethylformamide) and 0,5 M HBTU in DMF were prepared. The Tentagel R RAM resin (1,11 g) was allowed to expand in DMF for 30 minutes in the reaction vessel (illustrated below in fig. 9). The peptide elongation consisted of deprotection of preceding amino acid by treating the solution with 7 ml 20% piperidine in DMF for 2 minutes twice under rotation. After deprotection a washing step with 0,5 l DMF was carried out. The amino acids used for the DBZ- sequence were dissolved in 1,6 ml HBTU in DMF and 138,8 µl DIPEA
(Diisopropylethylamine) to a concentration of 0,8 mM (which was the determined molar abundance of 4 equivalents) before their addition to the reaction vessel. The
reaction time was set to 20 minutes except for proline (40 minutes). The reaction vessel was set to rotate during the amino acid coupling. The step reaction yield minimum was set to 99,4 % as indicated by the ninhydrin test. In case of lower yields, another mixture of amino acid, HBTU and DIPEA in a second coupling attempt. The reaction time was gradually increased when the first coupling yields were found too low for three amino acids in a row. The reaction time variation was from 20 minutes to 2 hours. After amino acid coupling the resin was washed with DMF which concluded the cycle and
deprotection for the next amino acid commenced. The first half of the two other
segments was synthesized as one segment, where the amount of reagents were the same as described above. For the second half of the segments, which was synthesized in two pots instead of one, half the amount of reagents was used with the exception of the washing and deprotection reagents. Another deviation from the amino acid coupling described above was the coupling of Ile, where 1,6 ml HATU in DMF was used instead of HBTU.
Figure 9: The reaction vessel used for peptide synthesis.
Ninhydrin tests
Three solutions for the ninhydrin test were prepared. The first, labelled solution A, consisted of 0,28 M ninhydrin in ethanol. The second, labelled solution B, was 0, 2 mM KCN (potassium cyanide) in pyridine. The third, solution C, consisted of 76 % phenol in ethanol. A small amount of the peptide was extracted after each amino acid coupling.
The sample was vacuum dried and weighed out at approximately 10 mg. The sample was placed in a test tube before addition of 2 drops solution A, 4 drops solution B and 2 drops solution C. During the solution addition a blank sample was prepared, containing everything except the resin adduct. The two tubes were placed in a heating block, at 100
°C for 5 minutes. Subsequently, each tube received 2,6 ml ethanol before measurement of the absorbance at 600 nm. This wavelength is different from what is generally employed (570 nm), and motivated by logistics (led to use of a fixed wavelength instrument). A conversion factor based on the ration A600/A570 was used to account for this deviation.
Peptide cleavage
Three solutions were prepared and named eluent A, eluent B and eluent A&B. Eluent A consisted of 99,95 % H2O and 0,05 % TFA (Trifluoroacetic acid). Eluent B consisted of 90% MeCN (Acetonitrile), 9,95 % H2O and 0,05 % TFA. Eluent A&B was a 50/50 mixture of eluent A and B. The cleavage was initiated by cleavage medium added in the following order; 0,5 ml H2O, 0,5 ml TIPS (triisopropylsilane) and 20 ml TFA. The cleavage medium was allowed to react for three hours under agitation. The peptides were transferred to separate Erlenmeyer flasks, and the funnels were washed with TFA.
The three beakers were labelled R-segment, W-segment and DBZ. The contents were subjected to nitrogen gas to remove TFA, and repeatedly weighed until no further weight change was observed. Subsequently, the peptides were transferred to 100 ml separation funnels and the beakers were washed with eluent A&B. Each beaker then received an addition of 30 ml diethyl ether and was agitated until separation was complete. The H2O phase was collected in a pre-weighed 200 ml round-bottom flask.
This procedure was carried out three times for each peptide residue. Each round-bottom flask was treated with nitrogen gas until the smell of diethyl ether had vanished. The three were placed in a rotation device in a shell freezer. To complete the freeze drying process the round-bottom flasks were placed to dry in the vacuum evaporator overnight.
The round-bottom flasks were weighed and crude yields calculated.
Peptide ligation
A preliminary test of ligation was carried out as described earlier (15). Phosphate buffer with a concentration of 200 mM and pH 7-7,2 was prepared. The ligation buffer was prepared using 100 ml of the 200 mM phosphate buffer with addition of guanidine to a final concentration of 6 M. An extracted volume of 20 ml of the guanidine solution received addition of MPAA (4-nitrophenyl chloroformate) to a concentration of 200 mM and TCEP (4-mercaptophenylacetic acid) to a concentration of 20 mM. The pH was kept at the interval 7-7,2 by addition of NaOH (sodium hydroxide). The ligation was carried out in a tinfoil-covered beaker overnight with magnetic stirring. The concentration of each peptide fraction (R-segment and Dbz-segment/ W-segment and Dbz-segment) was estimated to be 1 mM. The product was not analyzed further due to lack of time.
MS analysis
Samples of the crude products were desalted using a PD-10 gel filtration columns (GE healthcare), employing water: acetonitrile 30:70 (v/v) in a gravity protocol. The relevant fractions were pooled, freeze-dried, and redissolved in the same solvent. These samples were then used for MS injections.
Single trace injections of crude samples of the R, W, and Dbz segments, respectively were performed to get a first glimpse of the compositions of the product fractions, using an Agilent 1200 series HPLC system equipped with a single quadrupole mass
spectrometer and a DAD diode array multiple wavelength detector. Solvent used was 10 mM ammonium acetat, pH 7,0 (AmAc). Further parameters were as follows: flow rate:
0,1 ml/min, injection volume: 0,5 µl, sample concentration: 1 µg/ml. These steps were carried out by the supervisor
4 Results
The synthesis of the three AvBD 8 segments (W, R, Dbz) resulted in apparently reasonable yields. The crude yield of the W-segment (2621 Da) fraction was 137,6 mg (52,3%), since the theoretical yield was 262,1 mg. The R-segment with an expected molecular weight of 2591 Da and a theoretical yield of 259,1 mg and weighed 166,3 mg
which indicates a crude yield of 64,2 %. The Dbz-segment with an expected molecular weight of 2105 Da (including Dbz-linker) and a theoretical yield of 421 mg had a crude yield of 67,6 % which is 284,5 mg. The data is described below in Table 1.
Table 1
Molecular weights (Da), crude yields (mg), theoretical yields (mg) and crude yields (%) of the peptide segments.
Peptide Segment
Molecular weight (Da)
Crude yield (mg)
Theoretical yield (mg)
Crude yield (%)
Dbz-segment 2105 284,5 421 67,6
R-segment 2591 166,3 259,1 64,2
W-segment 2621 137,6 262,1 52,3
The ninhydrin test indicated a mean product yield of 89,1 % for the Dbz-segment, 92,7
% for the R-segment and 92,7 % for the W-segment. The yields are presented in detail below in table 1.2.
Table 1
Yield percentage of each ninhydrin test with the exception of proline couplings which were not measured.
Amino acid
Dbz- segment yield (%)
Amino acid R-segment yield (%)
Amino acid W-segment yield (%)
DBZ 99,4 S 99,9 S 99,9
H 97,7 V 99,8 V 99,8
D 99,4 V 99,7 V 99,7
T 99,6 T 99,8 T 99,8
S 99,6 R 99,8 R 99,8
C 99,8 C 99,9 C 99,9
T 99,4 C 99,7 C 99,7
G 99,7 I 99,8 I 99,9
G 99,7 G 99,4 G 99,8
A 99,4 Q 99,4 Q 99,4
Q 99,2 K 99,9 K 99,9
K 99,5 C 99,8 C 99,8
C 99,1 R 99,4 W 99,4
Q 99,7 G 99,4 G 99,4
V 99,6 F 99,4 F 99,4
A 99,4 A 99,4 A 99,4
N 99,2 R 99,4 R 99,4
N 99,8 T 99,5 T 99,5
A 99,3 N 99,4 N 99,4
J 99,5 J 99,5
Theoretical product yield (%)
Theoretical product yield (%)
Theoretical product yield (%)
89,1 92,7 92,7
MS analysis
The three trace MS injections of crude products (carried out by my supervisor) were merely to get an indication of their compositions. Evidently, chromatographic
purification should be performed for a proper assessment of product quality, but due to lack of time a quick, albeit inadequate, assement was better than none.
It appears that the Dbz segment fraction is composed by a multitude of peptide
fragments, and that the peaks emanating from the correct peptide mass (MW: 2105 Da;
m/z: 2106, 1053.5, 702.7, 527, 422) are present, if at all, in only minor amounts (Fig.
10).
Figure 10: MS trace of Dbz injection showing a multitude of major peaks.
The W-segment injection (Fig. 11) has a cleaner appearance, but although correct mass determinations are difficult from this coarse data, it appears that the major peaks do not emanate from fragments of the correct mass (MW: 2621 Da; mz: 2622, 1311.5, 874.3, 525.4). Observed peaks att 860 and 645 indicates a mass of 2577 Da, and another peak at 878 would correspond to 2631.
Figure 11: MS trace of W segment showing major peaks at 860 and 645.
The R segment injection (Fig. 12) shows major peaks at 1275, 852, 639 and 511, indicating the presence of a mass of around 2550 (2548, 2553, 2552, 2550,
respectively). The expected mass: 2591 thus differs by 41. Whereas the mass deviation appears too small to indicate an amino acid deletion, it matches the mass of minor structures, such as a carboxy group.
Figure 12: MS trace of R segment showing major peaks at 1275, 852, 639 and 511.
5 Discussion
The crude product yields were at an acceptable level seeing as product was continuously used to validate the coupling efficiency with the ninhydrin test. Further crude product yields in comparison to the ninhydrin test yields implicates that the R- and Dbz-segment have been treated rather uniformly since the yield reduction is approximately 28 % for both segments. The W-segment has a bigger yield reduction of approximately 40 %.
The reduction can be explained by the extracted amount of peptide product used to validate the coupling efficiency and due to the fact that a number of amino acids had to be double coupled when the yield percentage was too low (below 99,4 %). The problem with coupling efficiency also manifested itself in coupling time which had to be
adjusted over the course of time and peptides with coupling time of about 20 minutes had to be upgraded to times up until three hours. One explanation for the coupling efficiency can be a steric hindrance, which may occur due to entanglement of previously coupled amino acids. The MS spectras were inconclusive when determining whether the peptides were successfully synthesized or not, although preliminary interpretations are not encouraging. If correctly interpreted, apparently the synthesis has been
unsuccessful. This appears evident in the case of the dbz segment, exhibiting an
abundant mixture of masses. Further purification is however necessary and it is possible that this may change these intepretations.
One can speculate in error sources if the purification would fail to show adequate abundances of the correct structures. Certain amino acids are known to be difficult to couple in high yields, such as for instance Ile, but this was circumvented by the use of HATU for coupling (1). Coupling of histidine to the Dbz fragment had a relatively low yield (97,7%) but otherwise low coupling yelds could be overcome using double couplings and prolonged coupling times.. Another source of error could be the human factor, where, for instance, an addition of piperidine would lead to failed coupling of the following amino acid. However this is not likely since synthesis sheets were used and
carefully followed. A third possible source of error could be the condition of the reagents used where for instance an amino acid could be non-functioning, which could be an explanation for the extended coupling time.
For future studies the uncertainty of the reagents can be avoided by screen-testing the material by making small peptides and upon analysis confirm whether they work or not, and by MS analysis of individual amino acid reagents. Another method to avoid the uncertainty could be to continuously analyze fragments of the peptides during the synthesis, which was the initial thought. However due to the unavailability of a LC-MS machinery at the beginning of the synthesis this thought was withdrawn.
6 References
1. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev Immunol. 2003;3(9):710-20.
2. Lai Y, Gallo RL. AMPed up immunity: how antimicrobial peptides have multiple roles in immune defense. Trends Immunol. 2009;30(3):131-41.
3. van Dijk A, Veldhuizen EJ, Haagsman HP. Avian defensins. Vet Immunol Immunopathol. 2008;124(1-2):1-18.
4. Hellgren O, Sheldon BC, Buckling A. In vitro tests of natural allelic variation of innate immune genes (avian β-defensins) reveal functional differences in microbial inhibition. J Evol Biol. 2010;23(12):2726-30.
5. Zhu S, Gao B. Evolutionary origin of β-defensins. Dev Comp Immunol.
2013;39(1-2):79-84.
6. Klüver E, Schulz-Maronde S, Scheid S, Meyer B, Forssmann WG, Adermann K. Structure-activity relation of human beta-defensin 3: influence of disulfide bonds and cysteine substitution on antimicrobial activity and cytotoxicity.
Biochemistry. 2005;44(28):9804-16.
7. Ma D, Zhang K, Zhang M, Xin S, Liu X, Han Z, et al. Identification, expression and activity analyses of five novel duck beta-defensins. PLoS One.
2012;7(10):e47743.
8. Ma D, Lin L, Zhang K, Han Z, Shao Y, Liu X, et al. Three novel Anas platyrhynchos avian β-defensins, upregulated by duck hepatitis virus, with antibacterial and antiviral activities. Mol Immunol. 2011;49(1-2):84-96.
9. Chandrudu S, Simerska P, Toth I. Chemical methods for peptide and protein production. Molecules. 2013;18(4):4373-88.
10. Fischer E. Ueber die Hydrolyse der Proteïnstoffe. Chem Ztg. 1902;vol.
26(p. 939).
11. Bruckdorfer T, Marder O, Albericio F. From production of peptides in milligram amounts for research to multi-tons quantities for drugs of the future. Curr Pharm Biotechnol. 2004;5(1):29-43.
12. Hojo K, Hara A, Kitai H, Onishi M, Ichikawa H, Fukumori Y, et al.
Development of a method for environmentally friendly chemical peptide synthesis in water using water-dispersible amino acid nanoparticles. Chem Cent J. 2011;5:49.
13. Merrifield B. Bruce Merrifield - Nobel Lecture: Solid Phase Synthesis.
http://www.nobelprize.org/nobel_prizes/chemistry/laureates/1984/merrifield- lecture.html: Nobel Media AB [cited 2013 13/11].
14. Holme DJ, Peck H. Analytical biochemistry. 2. ed. Harlow: Longman;
1993. 507 s. p.
15. Gunasekera S, Aboye TL, Madian WA, El-Seedi HR, Göransson U.
Making Ends Meet: Microwave-Accelerated Synthesis of Cyclic and Disulfide Rich Proteins Via In Situ Thioesterification and Native Chemical Ligation. Int J Pept Res Ther. 2013;19(1):43-54.
16. Blanco-Canosa JB, Dawson PE. An efficient Fmoc-SPPS approach for the generation of thioester peptide precursors for use in native chemical ligation. Angew Chem Int Ed Engl. 2008;47(36):6851-5.
17. Snyder LR, Kirkland JJ, Dolan JW. Introduction to modern liquid chromatography. 3. ed. Oxford: Wiley-Blackwell; 2010. xli, 912 s. p.
18. Skoog DA, Holler FJ, Crouch SR. Principles of instrumental analysis. 6., [rev. and updated] ed. Belmont, CA: Thomson Brooks/Cole; 2007. xv, 1039 s. p.
19. Simonsen F, Lindegren R. Analysteknik : instrument och metoder. Lund:
Studentlitteratur; 2005. 375 s. p.
20. Freeze-drying. www.wikipedia.org; [cited 2013 10/12]; Available from:
http://en.wikipedia.org/wiki/Freeze-drying.
